EP2304641A1

EP2304641A1 - Method and apparatus for identifying a person using the typing behaviour of the latter and taking into account the local distribution of the keys of a keyboard

Info

Publication number: EP2304641A1
Application number: EP09765458A
Authority: EP
Inventors: Thomas Wölfl; Thomas Grundner
Original assignee: Psylock GmbH
Current assignee: Psylock GmbH
Priority date: 2008-06-19
Filing date: 2009-06-19
Publication date: 2011-04-06
Also published as: US20110185416A1; JP2011524580A; DE102008002544A1; EA201071325A1; CN102067149A; WO2009152822A1

Abstract

The present invention relates to a method and an apparatus for identifying a person using the typing behaviour of the latter, wherein the local position of the actuated keys (14) is taken into account when evaluating actuation of the keys.

Description

METHOD AND DEVICE FOR IDENTIFYING A PERSON BY MEANS OF YOUR TIP BEHAVIOR IN TAKING INTO ACCOUNT THE LOCAL DISTRIBUTION OF THE KEYS OF A KEYBOARD

BACKGROUND OF THE INVENTION

FIELD OF THE INVENTION

The present invention relates to a method and a device for identifying a person in which a keyboard is provided with a plurality of keys to which the person to be identified makes an input in which a plurality of keys are operated, so that due to the characteristic of the key operation identification of the Person takes.

STATE OF THE ART

From the prior art Identifϊzierungsverfahren and devices are known for people who use the typing behavior of the person to be identified on a keyboard. Examples are given in WO 98/06020 A2 and US Pat. No. 6,151,593.

Although these identification methods and devices are already achieving very good results, there is a continuing need to increase the security and reliability of such methods and devices and to improve the convenience by entering short typing sequences. These goals, which are actually incompatible because high levels of safety can usually be achieved with a long typing sequence, make it necessary to constantly think about improvements to these methods.

DISCLOSURE OF THE INVENTION

OBJECT OF THE INVENTION

It is therefore an object of the present invention to provide a method and a device for identifying a person on the basis of their typing behavior, which have the opposing goals of high security and reliability and convenient use, especially by short typing sequences, perform better than the methods known from the prior art.

TECHNICAL SOLUTION

This object is achieved by a method having the features of claim 1 and a device having the features of claim 13. Advantageous embodiments are the subject of the dependent claims.

The invention is based on the recognition that an improvement of the identification possibility and thus increase the identification security and reliability is given by the local position of the keys to be operated is included in the evaluation. As a result, further characteristics of the typing behavior of a user, which are due to the spatial distribution of the keys on a keyboard, can be utilized for the identification or influences when using different keyboards can be avoided.

The invention can be used in combination with all known methods and devices for identifying a user on the basis of his typing behavior, the prior art methods and devices usually having detection devices which record tipproh data upon actuation of the keyboard, which information about the actuated buttons and / or or the time of the operation and / or the duration and the nature of the operation. The type of operation can be on the one hand, the information about whether it is the pressing and / or releasing the button and / or with what pressure the button or a touch-sensitive typing field or the like is actuated.

In order to incorporate the local position of the actuated keys in the evaluation, the method may provide information about the distance of the keys from each other, the distance of groups of keys to each other, for example keypads such as the numeric keypad on a PC keyboard Distance of the keys or groups of keys to each other, for example along coordinate axes, as well as the coordinates of keys are provided in a coordinate system. Further information about the local position or distribution of the keys can be found in the information about the Order along certain lines, columns, rows, the arrangement in certain fields and in turn the position within the rows, columns or fields and / or neighborhood information to the individual keys include. Overall, a wide variety of information about the local position of the actuated buttons can be provided and used.

When evaluating the key operation, a variety of characteristic features can be determined from the raw tip data, such as the holding duration of a key, the transition period from releasing a key to press the next key, the transition period from releasing a key to release the next key, the transition duration Furthermore, the error frequency or frequency, for example, displayed by using the delete key or the tip rate, which provides information about how many words are written in a certain time, can also be used, wherein For example, the tip rate can be determined by the number of typed spaces. In addition, the attack frequency, ie the number of key presses per unit time, as well as other information regarding the frequency of use of certain function keys, the selection of certain keys that are provided on the keyboard repeatedly, such as the use of the left or right shift key , are used. Also special abnormalities, such as temporal or key-related values, which fall outside the usual framework, can be particularly used. For example, overhauls, d. H. the pressing of the next key before the preceding key is released can be used as a characteristic feature for the typing behavior of a person, whereby these overhauls can also take place repeatedly, ie with several keys in succession.

The consideration of the local position of the actuated keys can now take place in such a way that the local position of the keys, ie corresponding coordinates or distance values or the like, are set in relation to determined raw tip data and / or features so as to generate location-related features.

Thus, for example, by the relation of the distance of successive keys and a transition period, a transition speed can be generated as a location-related feature. With direction-related distance values, this can also lead to directional speeds as location-related features. Overall, distance characteristics and time characteristics can be used to determine speed characteristics as location-related features that can contribute to improved identification.

The features for multiple key presses, i. H. both the location-related features and the non-location-related features may be further processed, for example, to feature data items representing, for example, averages or totals or the like for particular determined data. Thus, for example, an average value can be formed as a function data feature for the holding duration of the keys for all keys. Moreover, it is also possible to group features into feature fields or vectors, where either similar or similar features can be grouped or summarized for different key presses, such as holding durations for all keystrokes detected on input, or different features of the same key press, such as hold durations Transition times, transition speeds, etc. In this case, corresponding function data characteristics, such as average holding durations, can also be taken over into the feature fields or vectors. Overall, a diverse treatment of raw tip data and features is possible.

By the method according to the invention it is possible to enable the identification of persons even with different keyboards, so that people who want to log into different computer systems in a specific system worldwide and need to be identified, different keyboards, as they are used worldwide come to use, since the different arrangement of the respective keys is taken into account by the method according to the invention.

In a corresponding device, in addition to a keyboard for inputting a corresponding typing sequence, a detection device for acquiring raw tip data and a storage unit for storing reference data and a comparison unit for comparing the reference data with data from the typing sequence entered in the access request must be provided, wherein the device has information about there must be a local distribution of keys on the keyboard or there must be an opportunity to gain access to this information. This can be done, for example, by must be entered into the device control, which keyboard is used. Alternatively, the device can determine itself by means of an automatic identification of the connected keyboard corresponding information, for example, from a central memory location, for example via remote data communication, the position data of the keys are queried to a keyboard type. This is easily possible, in particular, if an evaluation unit of the device is formed by programming design of a data processing unit, since then communication means of the data processing unit, such as network connections and the like, can be used. Accordingly, the detection device additionally or alternatively to the keyboard associated sensors can be realized in particular by a data processing program which detects the electrical signals of the keyboard or a typed field with appropriate operation.

Preferably, the identification device and in particular the evaluation unit and / or comparison unit is formed by a programmatically designed data processing device or integrated into this.

BRIEF DESCRIPTION OF THE FIGURES

Further advantages, characteristics and features of the present invention will become apparent in the following detailed description of exemplary embodiments. The drawings show this in a purely schematic way in

Figure 1 is a schematic representation of a device according to the invention with a keyboard for entering the typing sequence for identification; Figure 2 is a plan view of a keyboard for use with the present invention;

Figure 3 is a plan view of another embodiment of a keyboard for use with the present invention;

Figure 4 is a diagram illustrating the timing of a keyboard operation using the example of the keys for the letters x, y and z.

PREFERRED EMBODIMENTS 1 shows a schematic representation of a keyboard 1, which is operatively connected to a data processing device, such as a personal computer PC 2, wherein the data processing device 2, a display device 3, such as a monitor or screen can be arranged.

The keyboard 1 comprises a plurality of keys or keypads 4, which can be operated by pressing with a finger.

The keyboard 1 shown in FIG. 1 may, for example, be a restricted number block which is only used for access control for a room or the like.

As is apparent from Figures 2 and 3 with corresponding further examples of keyboards 10 and 100, the keyboards for use in the present invention may be of a different type. For example, FIG. 2 shows a keyboard commonly used with a personal computer PC. Accordingly, the keyboard 10 has a control and data line 11 for connection to a corresponding data processing device (not shown). The keyboard 10 also has a plurality of keys or keypads 14 arranged in two key blocks 12 and 13. For example, a common PC keyboard may be the so-called numeric keypad (keypad 13) and a keypad with keys that include the letters of the alphabet, function keys, etc. (keypad 12).

The keys 14 are arranged in the keyboard of the embodiment of FIG. 2 in columns 15 and rows 16 arranged perpendicular to one another, so that they extend along the axes of a Cartesian coordinate system x-y. Correspondingly, each key or each typing field 14 can be assigned an x-y coordinate, wherein due to the spatial extent of the key or typing field 14, the center or center point of a corresponding key 14 can be selected as the reference point. Alternatively, the keys may also be arranged in staggered rows or columns or in oblique rows or columns or in any desired shape.

For example, if the extent of a key 14 along the x or y coordinate is considered to be a unit of length, the key for the letter x indicates the xy coordinates (3, 4). while the key for the letter y has the xy coordinates (6, 3) and the key 14 for the letter z has the xy coordinates (7, 1). Similarly, all keys 14 of keypad 12 and keypad 13 can be assigned unique spatial locations.

From the spatial positions and relations of the keys 14 to each other can be determined. For example, the distance or the distance d ₃ between the keys for the letters x and y can be determined from the xy coordinates. For this purpose, first the distance of the keys for the letters x and y in the x-direction, which is denoted here by d _ϊ , can be determined, and then the distance of the keys for the letters x and y in the y-direction, here is denoted by d ₂ . d _ϊ in the example of FIG. 2 is 3, while the distance d ₂ in the y-direction is 1. The distance d ₃ results from the formula d ₃ ² = d] ² + d ₂ ² , so that in the present case the value for the distance to d ₃ =

Λ / 3 ² + 1 ² = * JΪÖ results. Alternatively, a relation of the keys 14 can also be determined separately for the x and y directions, that is, for example, distances in the x and y directions, or in other arbitrary directions.

In addition, relationships of groups of keys such as keypads 12 and 13 of the keyboard of FIG. 2 may also be determined. Thus, the distance d _{4 of} the key fields can be used. The keypads or groups of keys may be arbitrarily selected and are not limited to physically separate keypads such as the numeric keypad and the remaining keys of the keypad of FIG. Rather, keypads within an array of adjacent keys or keypads 14 may also be defined.

As can be seen from FIG. 3, any other arbitrarily shaped keyboard can also be used for the invention, in which case also the local distribution of the keys, for example using a Cartesian coordinate system, can be clearly identified. However, instead of a Cartesian coordinate system, another coordinate system, such as a polar coordinate system, may be used.

The keyboard 100, which in turn comprises a data and control line 111 to a data processing device, not shown, similar to the embodiment of Figure 2, a keypad 112 and a keypad 113, which are spatially separated by a Distance are separated. Keypads 112 and 113 include a plurality of keys or tine pads 114 which, however, are not arranged in straight and mutually perpendicular rows and columns, as in Figure 2, but in angularly spaced columns 115 and rows 116 which are curved. Accordingly, it is appropriate to use a polar coordinate system (φ, r) according to the radius of gyration to define the exact local distribution of the keys 114. Again, the individual positions of the keys can be described exactly and determine appropriate distances between the individual keys, the direct distance of the keys to each other and the directional distance along a specific axis or direction can be determined defined. In addition, the position of the individual keys can be defined by assigning them to certain rows and columns and specifying the position in the corresponding row or column by means of position numbers, for example.

In addition, in the embodiment of FIG. 3, as shown in FIG. 2, a distance d ₄ between the keypads 12 and 13 can also be determined and used to evaluate the typing behavior.

FIG. 4 shows in a flow chart along a time axis t the sequence of a typing sequence with the typing sequence xyzxy in which the corresponding keys for the letters x, y and z are pressed in the corresponding sequence. As can be seen from the diagram of FIG. 4, an actuation of the keys or tip pads 4, 14, 114 is carried out at the times a to j. At time a, the key for the letter x is pressed, while at time c, the key x is released. At time b, the key y is pressed and released at time e. The key z is depressed at time d and released at time f, while the key x is pressed again at time g and released at time h. Finally, the key y is pressed a second time at time i and released again at time j.

By pressing the individual keys, features such as the duration or the transition period can now be determined.

So z. B. the holding time t] for the key x from the difference of the times a and c. The transition time t ₂ for the push transition from letter x to y results from the difference between the time a, to which the x key is pressed, and the time b, too the key y is pressed while z. B. the release transition time t ₃ for the key sequence xy by releasing the key x at time c and the release of the key y at time e is defined.

In addition, by the detection device, for example, the pressure p can be determined, with which the keys 4, 14, 114 are acted upon by the user. This is likewise illustrated in FIG. 4, whereby here it is indicated purely schematically that there is a constant pressure between pressing and releasing. However, the pressure can also have a certain time course, which can also be used as a characteristic for the typing behavior. As the diagram of Figure 4 shows, the key x is subjected to a lower pressure than the key y or the key z, in turn, the key y is operated with a lower pressure than the key z.

By the detection device of a device according to the invention, which can be realized, for example, in a data processing device 2 by a corresponding data processing program in such a way that values determined by sensors on the keyboard are detected, information as to which buttons are actuated at what time is determined and stored , This information represents the raw tip data. From the raw tip data, corresponding features, such as transition durations t _2, t ₃ or holding duration t _1? be determined, which in turn can be further processed for all key presses or for specific keys to average values (function data characteristics). Similarly, many different features, such as transition durations, transition speeds or holding durations, or like features, such as holding durations for various specific keys, can be summarized into feature fields or vectors. The data determined in this way, that is to say all raw tip data as well as the characteristics derived therefrom, feature combinations and feature vectors, can be used to identify a person who makes a key input, wherein the corresponding data, which are determined during access control, are compared with reference data which are deposited for a specific user. The comparison can be made in a variety of ways by a direct pattern comparison or by statistical or other mathematical methods. It is essential that in the present invention, the position of the actuated keys is included, so that an improvement in the characterization of the typing behavior is possible. This results from the fact that the typing behavior is influenced by the local arrangement of the various keys, it being understood that, with consecutive keys that are close to each other, the transition durations are much shorter than with widely spaced keys.

For example, the key sequence, as shown in the diagram of FIG. 4, shows that with a key sequence xyz the subsequent keys are pressed much faster so that overhauls occur; H. the key y is already pressed before the key x is released and the key z is already pressed before the key y is released, while in the key sequence zx, in which there is a large distance between the successive keys, no overhaul takes place and the transition period accordingly is long.

By using position-related key information, corresponding speeds can also be determined, for example for the transition period t ₂ from the transition of the key x to the key y by dividing the distance d ₃ by the transition duration t ₂ . Directional velocities can also be determined, such as transition velocities in the x-direction or in the y-direction. For this purpose, the distances in the x-direction d] or in the y-direction d _{2 must be divided} by the corresponding transition duration t ₂ .

This transition speed can also be referred to as the Manhattan Speed as the keyboard is compared to the block layout in Manhattan and the shift from one key to another can be considered moving from one block in Manhattan to another.

Although the present invention has been described in detail with reference to the accompanying embodiments, it will be understood by those skilled in the art that the invention is not limited to these embodiments, but rather variations in the form of different combinations of individual features presented as well as omission of individual features, without departing from the scope of the appended claims. Overall, any combination of the features presented is claimed for the present invention.

Claims

claims

A method of identifying a person in which a keyboard (1, 10, 100) having a plurality of keys (4, 14, 114) is provided to which the person to be identified makes an input in which a plurality of keys are operated, on the basis of the characteristic of the key operation is an identification of the person, characterized in that in the evaluation of the key operation, the local position of the actuated keys is taken into account.

2. The method according to claim 1, characterized in that a detection device is provided which detects upon actuation of the keyboard raw tip data, the information about the actuated buttons (4,14,114) and / or the time and / or the time duration and / or include the type of operation, wherein the type of operation includes pressing and / or releasing the button and / or the pressing force on the button.

3. The method according to claim 1 or 2, characterized in that for evaluating the key operation at least one information from a group of information is provided, wherein the group information about the distance of the

Keys to each other, the distance of groups of keys to each other, the directional distance of the keys or groups of keys to each other, the coordinates of the keys in a coordinate system, the arrangement along lines or in boxes (12,13; 112,113), the arrangement in rows and / or columns and the location therein.

4. The method according to any one of the preceding claims, characterized in that in the evaluation of the key operation from the raw tip data for typing behavior at least one characteristic feature is determined from the group, the duration of a key, the transition period from releasing a key to press the next Key, the transition period from releasing a key to release the next key, the transition period from pressing a key to pressing the next key, the error frequency or frequency, the tip rate, the attack frequency, single or multiple overhauls, the frequency of use of certain function keys, the Selection from alternative keys, selection of the left or right shift key and temporal or key-related abnormalities, which are distinguished by exceptional values from the determined data.

5. The method according to any one of the preceding claims, characterized in that in the evaluation, the local position of the keys (4,14,114) is set in relation to raw tip data and / or features, so that location-related features are generated.

6. The method according to any one of the preceding claims, characterized in that in the evaluation of the distance of successive keys and a transition period of successive keys, a transition speed is generated as a location-related feature.

7. The method according to any one of the preceding claims, characterized in that direction-related speeds are determined as location-related features in the evaluation of distance values.

8. The method according to any one of the preceding claims, characterized in that speed characteristics are determined as location-related features in the evaluation with distance values and time characteristics.

9. The method according to any one of the preceding claims, characterized in that the features for a plurality of key presses to function data features comprising averages, sums or the like summarized or grouped to feature fields or vectors.

10. The method according to any one of the preceding claims, characterized in that several different features of the same key operation are grouped into feature fields or vectors.

11. The method according to any one of the preceding claims, characterized in that the identification by comparing reference data of the person to be identified with Access data of the person to be identified, which has generated the person to be identified during keyboard input.

12. The method according to any one of the preceding claims, characterized in that the inputs of the person to be identified can be done on different keyboards.

13. A device for identifying a person with a keyboard (1,10,100) having a plurality of keys (4,14,114) for inputting a typing sequence, a detection device for detecting raw tip data in the operation of the keyboard, a memory unit for storing reference data of one or a plurality of users, a comparison unit for comparing the reference data with the data obtained when accessing a person to be identified, characterized in that the device comprises information or access to information about the local distribution of the keys on the keyboard.

14. The device according to claim 13, characterized in that the device comprises an evaluation unit for processing the raw tip data in features for characterizing the typing behavior during the input, wherein the evaluation unit and / or comparison unit are formed by one, preferably the same data processing unit.

15. The apparatus of claim 13 or 14, characterized in that the information about the local distribution of the keys on a keyboard as stored information in a memory unit of the device or are available via data communication.

16. Device according to one of claims 13 to 15, characterized in that the device is prepared so that the method according to one of claims 1 to 12 is realized by the operation of a data processing program.